Orbital Remote Sensing for Carbon Monitoring in Tropical Forests: Accuracy and Potential Applications in the Atlantic Forest

• Grant number: 24/17429-4
• Support Opportunities: Scholarships in Brazil - Doctorate (Direct)
• Start date: April 01, 2025
• End date: March 31, 2029
• Field of knowledge: Biological Sciences - Ecology - Ecosystems Ecology
• Principal Investigator: Paulo Guilherme Molin
• Grantee: Thiago Almeida Bueno
• Host Institution: Escola Superior de Agricultura Luiz de Queiroz (ESALQ). Universidade de São Paulo (USP). Piracicaba , SP, Brazil
• Associated research grant: 21/11940-0 - Restoration of native vegetation in the Atlantic Forest through the strategic combination of mandatory measures and voluntary commitments - CCD-EMA, AP.CCD

Abstract

Tropical forests play a vital role in regulating the global climate and conserving biodiversity, serving as important carbon reservoirs essential for mitigating climate change. However, the persistent degradation and fragmentation of these forests continue to threaten their capacity for carbon sequestration and storage, highlighting the need for effective methods to frequently and widely monitor their structure, in order to enhance additional forest restoration initiatives. In this context, an increasing number of orbital remote sensing (SRorb) products for the global monitoring of structural parameters have become available, though their performance on smaller scales and within specific phytogeographic domains remains uncertain. This project aims to assess the sensitivity, in terms of accuracy, of the most promising SRorb products for measuring carbon density in tropical forests, as well as evaluate the potential of these products in generating reliable carbon increment curves and being used to detect and quantify edge effects on forest carbon. The research will focus on the Atlantic Forest biome and will employ a combination of extensive and robust field data, cutting-edge technologies such as orbital LiDAR, and advanced techniques using Artificial Intelligence (AI), including Convolutional Neural Networks (CNN), Random Forest (RF), and Self-Supervised Learning (SSL). The study will be conducted across the main forest typologies of the Atlantic Forest (Rainforests and Seasonal Forests). The results of this work have the potential to be expanded to other biomes and to foster tools and actions for public policies, such as the Atlantic Forest Law and climate action plans.